Modelling Physical and Chemical Properties of Aerosols: A Challenge for Air Quality and Climate Change Research

Transcription

1 Modelling Physical and Chemical Properties of Aerosols: A Challenge for Air Quality and Climate Change Research Mihaela Mircea Institute of Atmospheric Sciences and Climate National Research Council Bologna, Italy

2 What is aerosol?

3 Aerosol sources

4 Aerosol processes

5 Aerosol models Ae ros ol p roc ess es s e c r u so l o s ro e A Aerosol properties

6 Why is atmospheric aerosol so important? Aerosol affects: -air quality and health; -climate changes; and is used in: -geoengineering experiments

7 M2M for air quality and climate change Physical theories about: -how air move -how water behave -how radiation transports energy. and physical and chemical theories about: -how gas and aerosol are transported, transformed and interact between them and with water and radiation. The earth s atmosphere is broken down into pieces-cells- and then it is computed what happens in each cell based on these physical theories. Cells exchange energy and matter with each other based on he physics of the transfer of matter and energy. The net results is a simulation of what the actual atmosphere might do. Air quality and climate change have many overlapping and inter-linked problems.

8 Common issue between air quality and climate change with respect to aerosols at process level in model development, application and validation in measurement and monitoring strategies (ACCENT Report )

9 Process level: Aerosol emissions Aerosol emissions have to be included air quality (AQM) and climate models (CM) as process emission. Natural sources of aerosol depend on atmospheric conditions (wind speed, temperature, solar radiation, relative humidity, etc), on surface type (soil, vegetation) and status (temperature, water content, etc.). Both emission inventories and processes have to contain information not only on total mass or number of aerosol particles emitted, but on their size, chemical composition, state of mixing, solubility, etc. Both emission inventories and processes have to describe the aerosol sources with much better temporal and spatial accuracy. Process emissions implemented in AQMs&CMs: -dust aerosol -sea-salt aerosol -biogenic VOC (aerosol precursors) Process emissions under implementation in AQMs&CMs: -primary organic marine aerosol

10 A combined organic-inorganic submicron sea spray source function for large-scale models Requires only wind speed and surface ocean chlorophyll-a concentration as input parameters. (O Dowd et al., GRL 2008)

11 The combined source function was implemented in the REMOTE regional climate model and sea-spray fields are predicted over the North East Atlantic

12 Process level: SOA formation Formation of low-volatility (semivolatile and possibly nonvolatile) compounds is governed by a complex series of reactions of a large number of organic species. Experimental characterization and theoretical description of SOA formation presents a substantial challenge. SOA models, currently based on laboratory measurements, underestimate SOA production for two reasons: uncertainties in current treatments of SOA formation in models exist additional classes of SOA precursors that have yet to be studied in the laboratory or included in models. (Kroll and Seinfeld, 2008) - two-product model or volatility basis set -the NOx dependence of SOA yields - the effects of specific reaction conditions (RH, acidity, etc.) -the effects of multigenerational chemistry -cloud processing -higher-volatility precursors (isoprene, acetylene), lower-volatility species intermediate volatility and semivolatile organics, such as those in diesel exhaust. -a number of as-yet unidentified SOA precursors, with a wide range of volatilities.

13 Air Pollution in Pittsburgh July 2, 2001 PM2.5=4 µg m-3 July 18, 2001 PM2.5=45 µg m-3 (from Spyros Pandis)

14 Process level: Aerosol water uptake.humidifcation mainly in the boundary layer. is source of inter-model variability.and increases the range of modelled direct radiative forcing (RF) (IPCC, 2007) Aerosol water uptake aerosol optical properties aerosol sedimentation and dry deposition Aerosol physical and chemical properties aerosol wet removal Need to know how aerosol water uptake depends on: size composition surface tension solubility mixing state shape gas condensation/dissolution

15 Process level: Aerosol interactions with clouds and precipitations Need to know: the relationship between aerosol size/composition and drop/ice size distributions; the relationship between aerosol characteristics and cloud dynamics; -0.7 W/m W/m2 -the aerosol effects on the surface energy budget and, consequently, on convection, evaporation and precipitation and on biogeochemistry; Glaciation effect Thermodynamic effect effect

16 Process level: Aerosol wet removal Removal of aerosol particles by hydrometeors is subect to different processes: -in cloud (rainout) and -below cloud (washout) and its magnitude depends on the microphysical characteristics of both aerosol and hydrometeors. (Textor et al., 2006) Outstanding research is needed: -for understanding and parameterize the effect of particle size and composition on wet scavenging coefficient; -for understanding the processes contributing to below-cloud wet removal (the effect of cloud entrainment for example); -for understanding and parameterize the aerosol scavenging as a function of hydrometeor type

17 Process level: Aerosol nucleation Nucleation mechanisms: condensation of a binary mixture of sulphuric acid and water; ternary nucleation of sulphuric acid, water and a third molecule, most likely ammonia; ion induced nucleation; secondary aerosol formation involving condensationorganic of low- or non-volatile organic compounds; homogeneous nucleation of iodine oxides. In addition, atmospheric particle formation events are significantly affected by environmental factors, such as temperature, humidity and the surface area of pre-existing particles. (Hamed et al., 2007) Current problems in understanding particle formation and growth: -inability to detect particles smaller than 3 nm -inability to know in real time, in situ, composition of particles smaller than 200nm

18 Process level: Aerosol heterogeneous and aqueous chemistry Heterogeneous chemistry: -clear evidences on gas scavenging (O3, HNO3, etc) via non-reactive processes and reactive heterogeneous chemistry on aerosol particles -nitrate formation still a problem (size dependence, role of particles such as dust, sea-salt, etc) -role of organics for aging (oxidation processes, oligomerization processes, etc.) Aqueous chemistry: -clear evidence of change in chemical composition due to sulfate oxidation, etc Both these processes impact on particle composition and thus, on their further interactions with radiation and water. Heterogeneous processes in troposphere are considerably less well understood than gas-phase chemistry for two reasons: detection of the effects of heterogeneous reactions are not easily observable because of the rapid mixing times in the troposphere and its chemical heterogeneity. the complexity that is observed in the chemical composition of the aerosol particles.

19 Process level: Aerosol dry deposition Dry deposition velocity of aerosol is quantified using resistance analogy approach: -the aerodynamic resistance to transfer ra = f ( U/u2,L) -the resistance to transfer across the quasi-laminar surface layer rb = f ( 1/u, D) -the resistance to surface uptake rc = f(surface structure, presence of water or films). -gravitational settling velocity vg -height z -friction velocity u -horizontal wind speed U -Monin-Obukhov length L -molecular diffusivity of C in air Outstanding research needed: -on assessing the magnitude and spatial extent of edge effects and landscape heterogeneity; -on understanding and parameterization of particle size and composition effect on dry particle fluxes; -on understanding the processes contributing to deposition efficiency (electroand thermophoresis for example); -on understanding and parameterization of resuspension, saltation or other particle formation processes at or close to canopies aerosol dry deposition or particle atmosphere surface exchange?

20 Process level: Aerosol-radiation interactions Dust-radiation interactions improve weather forecast (Perez et al., 2006) Inadequate understanding of optical/radiative properties of aerosol. (IPCC,2007)

21 BOLCHEM Flow Chart Aerosol model M7 aerosol optical properties, Meteorological cloudcondensationnuclei Model (BOLAM) Transport & diffusion Winds, T, P, q, Clouds, RadiativeFluxes Heterogeneous chemistry Gas Chemistry (SAPRC90/CBIV) Emissions gas&aerosol Dry and wet removal gas&aerosol

22 M7: size-resolved aerosol microphysical model

23 BOLCHEM: dry deposition and sedimentation Dry deposition velocity of aerosol particles Where fsnow, fbs, fwat, fveg, fws and fice are the surface fractions of snow, bare soil, water, vegetation, wet skin and ice, respectively. Simple upwind sedimentation scheme (Ganzeveld et al., 1995,1998, 2006)

24 BOLCHEM: dust model Dust emission depends on: Surface wind speed Vegetation type and cover Preferential source areas Soil particle granulometry (soil texture) Soil moisture, snow cover (Tegen et al., 2002; Tegen et al., 2004)

25 BOLCHEM-dust simulations Saharan Dust over Italy: July 16, 2003

26 BOLCHEM-dust simulations S1 S2 Vertical distribution of dust concentrations (µg/m3) at Etna, from 14 to 19 July, for the cases S1 and S2. The results are in agreement with Tafuro et al. (2006), the dust layers are located below 6 km (ca. level 19). The simulations S1 and S2 were carried out with threshold wind velocity lowered by a factor of 0.50 and 0.75 with respect to the values calculated according with Marticorena and Bergametti (1995).

27 BOLCHEM: sea salt model Source emissions model for seasalt: 1) Moanhan et al. (1986) adapted for M7 by P. Stier 2) Gong (2003) adapted for M7 by E. Viganti 16 July 2003, 12 UTC sea salt model 2 These results are similar to the results of Langmann et al. (2008) for June 2003.

28 BOLCHEM: aerosol wet removal The aerosol is removed: -with different scavenging coefficients for convective and stratiform precipitations -with different scavenging coefficients for in-cloud and below-cloud in case of stratiform precipitations. The scavenging coefficient is computed as a function of rain intensity. The amount of aerosol removed in a grid box is also a function of the cloud fraction contained in it. The present implementation considers only the removal by liquid precipitations. The scavenging coefficients are size and compound dependent. (Mircea et al., 2000)

29 BOLCHEM: aerosol wet removal simulations Stratiform precipitation: 26 September 2007 at 19:00 precipitation aerosol without wet removal aerosol with wet removal Convective precipitations: 12 June 2007 at 16:00 precipitation aerosol removal without wet aerosol removal without wet

30 Convective preci eestratiform preci eeaverage aerosol removed BOLCHEM: aerosol wet removal simulations average aerosol removed eebelow- and i eebel ee e

31 BOLCHEM: anthropogenic aerosol

32 BOLCHEM: gas phase photochemistry

35 BOLCHEM esiste per lo sforzo diretto di: (in ordine alfabetico) Andrea Armaroli (supporto informatico) Rita Cesari Massimo D Isidoro Alberto Maurizi Mihaela Mircea Claudia Pizzigalli Francesca Monti Francesco Tampieri lo sforzo indiretto di: (in ordine alfabetico) Stefano Decesari Maria Cristina Facchini Sandro Fuzzi E perché esiste BOLAM: Andrea Buzzi e Piero Malguzzi

36 ACKNOWLEDGEMENTS This work is conducted in the frame of ACCENT, GEMS and CITYZEN EC proects, Italian MIUR proect AEROCLOUDS, and was also supported by the Italian Ministry of Environment through the Program Italy-USA Cooperation on Science and Technology of Climate Change.

37